Table 1

Absolute and relative clinical event risks

Type of clinical eventPatients (%; 95% CI)*Patients (%; 95% CI)* Relative event risk; 95% CI*
H-GPI arm, 1 yearH-GPI arm, 3 yearsBivalirudin arm, 1 year Bivalirudin versus H-GPIBivalirudin arm, 3 years Bivalirudin versus H-GPI
HORIZONS-AMI trial
 Non-CABG-related major bleeding
  • 165 (9.2; 7.9 to 10.6)

185 (10.3; 8.9 to 11.8)
  • 103 (5.7; 4.7 to 6.9)

  • 0.63; 0.49 to 0.79

  • 121 (6.7; 5.6 to 8.0)

  • 0.66; 0.53 to 0.82

 Non-CABG-related major access site bleeding64 (3.6; 2.7 to 4.5)67 (3.7; 2.9 to 4.7)
  • 34 (1.9; 1.3 to 2.6)

  • 0.53; 0.35 to 0.80

  • 39 (2.2; 1.5 to 3.0)

  • 0.58; 0.39 to 0.86

 Non-CABG-related major non-access site bleeding101 (5.6; 4.6 to 6.8)118 (6.5; 5.4 to 7.8)
  • 69 (3.8; 3.0 to 4.8)

  • 0.68; 0.51 to 0.92

  • 82 (4.6; 3.6 to 5.6)

  • 0.70; 0.53 to 0.92

 Non-CABG-related minor bleeding256 (14.2; 12.6 to 15.9)262 (14.5; 12.9 to 16.3)
  • 137 (7.6; 6.4 to 8.9)

  • 0.54; 0.44 to 0.65

  • 146 (8.1; 6.9 to 9.5)

  • 0.56; 0.46 to 0.68

 Ischaemic stroke18 (1.0; 0.6 to 1.6)31 (1.7; 1.2 to 2.4)
  • 19 (1.1; 0.6 to 1.6)

  • 1.06; 0.56 to 2.01

  • 27 (1.5; 1.0 to 2.2)

  • 0.87; 0.52 to 1.46

 Repeat myocardial infarction76 (4.2; 3.3 to 5.3)135 (7.5; 6.3 to 8.8)
  • 62 (3.4; 2.7 to 4.4)

  • 0.82; 0.59 to 1.14

  • 105 (5.8; 4.8 to 7.0)

  • 0.78; 0.61 to 0.97

 Repeat revascularisation155 (8.6; 7.3 to 10.0)302 (16.8; 15.1 to 18.6)
  • 174 (9.7; 8.3 to 11.1)

  • 1.12; 0.91 to 1.38

  • 337 (18.7; 16.9 to 20.6)

  • 1.12; 0.97 to 1.29

 Death86 (4.8; 3.8 to 5.9)134 (7.4; 6.3 to 8.7)
  • 61 (3.4; 2.6 to 4.3)

  • 0.71; 0.52 to 0.98

  • 102 (5.7; 4.6 to 6.8)

  • 0.76; 0.59 to 0.98

Estimates for 42.5% radial arterial access use as assumed in the base case analysis§
 Non-CABG-related major bleeding138 (7.7; 6.5 to 9.0)157 (8.7; 7.5 to 10.1)
  • 89 (4.9; 4.0 to 6.0)

  • 0.64; 0.50 to 0.82

  • 104 (5.8; 4.7 to 7.0)

  • 0.67; 0.53 to 0.83

 Non-CABG-related major access site bleeding37 (2.1; 1.4 to 2.8)39 (2.2; 1.5 to 2.9)
  • 20 (1.1; 0.7 to 1.7)

  • 0.53 (0.32 to 0.93)

  • 22 (1.2; 0.8 to 1.8)

  • 0.58 (0.34 to 0.95)

 Non-CABG-related minor bleeding147 (8.2; 6.9 to 9.5)151 (8.4; 7.1 to 9.8)
  • 79 (4.4; 3.5 to 5.4)

  • 0.54; 0.44 to 0.66

  • 84 (4.7; 3.7 to 5.7)

  • 0.56; 0.45 to 0.69

  • * Based on Kaplan–Meier estimates of event numbers for the HORIZONS-AMI intention-to-treat population. H-GPI arm, N=1802; bivalirudin arm, N=1800. Limits of 95% CIs were used as ranges of variation in sensitivity analysis. In probabilistic sensitivity analysis, absolute (comparator strategy) event risks were represented by β distributions and relative event risks by lognormal distributions.

  • The HORIZONS-AMI definition of major bleeding included any intracranial bleeding; intraocular bleeding; retroperitoneal bleeding; access site haemorrhage requiring surgery; haematoma ≥5 cm; reduction in haemoglobin of ≥4 g/dl without overt source; reduction in haemoglobin of ≥3 g/dl with overt source; re-operation for bleeding; use of any blood transfusion.5

  • Patients with a non-CABG major bleeding were classified as ‘access site’ if they had a reported arterial access site or retroperitoneal bleeding, but no other major bleeding. The remainder of patients with a major bleeding were classified as ‘non-access site’. An equivalent definition was recently used by Verheugt et al.14

  • § Estimates based on assumption of no access site bleedings in patients with radial arterial access (see methods, section on clinical event risks).

  • CABG, coronary artery bypass graft; H-GPI, heparin and glycoprotein IIb/IIIa inhibitor; HORIZONS-AMI, Harmonizing Outcomes with Revascularisation and Stents in Acute Myocardial Infarction.